Killer immunoglobulin-like receptors (KIR) are a family of receptors that, along with C-type lectin receptors (CD94-NKG2), are used by human NK cells and T-lymphocyte subsets to specifically recognize MHC class I molecules. Certain inhibitory and activating KIR have highly similar extracellular domains and are recognised by the same monoclonal antibody, e.g. KIR2DL1 and KIR2DS1 are both recognised by EB6, and 2DL2 and 2DS2 by GL183. Three criteria (number of extracellular Ig-like domains (domains D0, D1, D2), cytoplasmic tail length, and sequence analogy) have been used to categorise the KIR proteins into 13 groups, namely KIR3DL1-2, KIR3DS1, KIR2DL1-5, and KIR2DS1-5. The nomenclature 2D for 2 domains or 3D for 3 domains gives the number of Ig-like domains; receptors with either long or short cytoplasmic domains are further classified as L or S. (Pascal V. et al., 2007 J. Immunol. 179:1625-1633) The inhibitory receptors possess long (L) cytoplasmic tails (i.e., KIR2DL or KIR3DL) containing a canonical ITIM that becomes tyrosine phosphorylated upon KIR engagement of their HLA class I ligands. The phosphorylated ITIM recruits the Src homology 2 domain containing protein tyrosine phosphatases Src homology 2 domain-containing phosphatase 1 and/or Src homology 2 domain-containing phosphatase 2, which dephosphorylate cellular substrates, thus aborting the NK activation signal, i.e., sparing target cells with appropriate self-MHC class I expression. Receptors with short (S) cytoplasmic tails lack ITIMs (i.e., KIR2DS or KIR3DS). These activating KIR contain a charged residue within their transmembrane domain facilitating interaction with the signaling chain KARAP/DAP12. Engagement of the KIR2DS family of receptors has been shown to lead to a cascade of KARAP/DAP12-mediated signaling events culminating in increased NK cell cytolytic activity and the production of proinflammatory cytokines such as IFN-γ (Pascal et al. 2007 J. Immunol. 179: 1625-1633). Mature NK cells are predicted to acquire at least one inhibitory receptor specific for a self-MHC class I molecule, which generally functionally prevails over potentially auto-reactive activating molecules. It is proposed that the response of NK cells represents the integrated outcome of both activating and inhibitory signalling by KIR and other receptors.
X-ray crystallographic analysis has provided high-resolution images of KIR2DL2 bound to HLA-Cw3 and of KIR2DL1 bound to HLA-Cw4 (Boyington, et al. (2000) Nature. 405:537-543). In both complexes loops from the Ig-like domains D1 and D2 of KIR2D are involved in binding to HLA molecules. In comparison to the interactions of KIR2D with HLA-C, little is known of the interaction between KIR3D and their ligands HLA-B or HLA-A. The KIR2D genes encoding HLA-C receptors form part of a larger group of KIR called lineage III KIR and all KIR2D genes of lineage III contain a pseudoexon encoding a D0 domain that is not incorporated into mature RNA. KIR2D genes are thus believed to have evolved from KIR3D genes. Human KIR specific for HLA-A and B form part of KIR lineage II which is comprised solely of KIR3D which all comprise a D0 domain. The D0 domain is the most N-terminal Ig-like domain in KIR3D proteins. While some reports suggest that the D0 domain is not involved in ligand-induced signalling (e.g., Snyder et al. (1999) Proc. Natl. Acad. Sci. USA. 96:3864-3869), others have proposed models where the D0 domain does not participate in ligand binding it may have an enhancing role in signaling. Khakoo et al. (2002, J. Exp. Med. 196(7):911-921) reported that various point mutations in the D1 and D2 domains of KIR3DL1, but none of 15 different point mutations in D0, abrogated KIR3DL1 binding to Bw4+ HLA-B.
It has been reported that several malignancies, autoimmune or inflammatory disorders involve CD4+ T cells that express KIR3D receptors. However, the functional role of KIR in T cells is largely unknown, and KIR-mediated signalling has been reported only in CD8+T and NK cells, in which case KIR have been involved in regulating effector cell cytotoxicity. The little knowledge of KIR signalling in CD4+ T cells has been limited to activatory KIR (e.g. KIRDS polypeptides), and these have been reported only to have a co-activatory role, rather than the true activatory role in NK cells (see e.g. Namekawa 2000, supra). The lack of true activatory function for KIRDS was reported to be due to missing “DAP12” signalling adaptors in T cells (Snyder et al. (2003) J. Exp. Med. 197(4):437-49).
The existence of numerous anti-KIR antibodies has been reported in the scientific literature. Shin et al. (1999) Hybridoma 18(6): 521-527 for example report a study of KIR antibodies. The majority of the antibodies that bound the KIR did not appear to inhibit the signal transduction mediated by the KIR and were therefore non-functional. Watzl C. et al. (2000) Tissue Antigens; 56: 240-247, identified “Lig1” antibodies that bind a common epitope on KIR2D receptors (except for KIR2DL4) but did not bind any KIR3D polypeptides. However, the Watzl et al. antibodies were not functional in their ability to block KIR-ligand mediated inhibition of NK cell cytotoxicity, nor did they inhibit binding of KIR-Ig fusion proteins to MHC class I-expressing cells. Other publications cite the existence of antibodies reactive against various KIR3D polypeptides. None of these antibodies are reported to distinguish KIR3D polypeptides from KIR2D polypeptides by binding all KIR3D (i.e. KIR3DL1, KIR3DL2 and KIR3DS1) yet no KIR2D polypeptides. Two anti-KIR3DL2 antibodies have been reported: Q241 and Q66 (Pende, et al. (1996) J Exp Med 184:505-518). These two antibodies are of the IgM isotype and will be expected to have low ADCC activity, or if their variable regions were placed in the context of a bivalent IgG type antibody, their affinity would generally decrease to the extent that significant ADCC induction would be precluded. The existence of a further KIR3DL2 antibody referred to only by the name of the “AZ158” cells producing it was reported (Parolini, S., et al. (2002) in Leucocyte Typing VII. D. Mason, editor. Oxford University Press, Oxford. 415-417). Several antibodies have been reported to bind the monomeric KIR3DL1 but not dimeric p140-KIR3DL2 (e.g. clone Z27 from the A. Morena group and DX9 from the L. Lanier group, both available from Becton Dickinson). While KIR3DL1 and KIR3DS1 share high sequence identity (KIR3DS1 is an activating form of the KIRDL1 gene), KIR3DL2 and KIR3DL1 differ in that KIR3DL2 are dimeric and share lower amino acid identity with KIR3DL1, including within the ECD (86% identity). KIR3DL1 recognize the MHC class I molecules HLA-B while KIR3DL2 recognize HLA-A.
Despite a great deal of research into the KIR family, and despite the existence of research reagents binding to various KIR, the role of certain KIR such as KIR3D polypeptides remained to be elucidated. There therefore remained a need for improving anti-KIR based therapies.